def visualize_age_relations(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "age_relations", dbutils)
display_positive = dataframe_pd["positive"]
display_negative = dataframe_pd["negative"]
display_age = dataframe_pd["age"]
rec_age_fig = make_subplots(
rows=1,
cols=2,
subplot_titles=("Positive test/age coefficient correlation",
"Negative test/age coefficient correlation"))
rec_age_fig.add_trace(go.Box(x=display_positive,
y=display_age,
name="Positive",
marker_color=color_scheme.color_300),
row=1,
col=1)
rec_age_fig.add_trace(go.Box(x=display_negative,
y=display_age,
name="Negative",
marker_color=color_scheme.color_700),
row=1,
col=2)
rec_age_fig.update_traces(boxpoints='all')
rec_age_fig.update_layout(
title_text="Subplots of age in relation a positive/negative test result"
)
rec_age_fig.show()
return None
def visualize_confirmed_cases_europe(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "confirmed_cases_europe", dbutils)
fig = px.choropleth(dataframe_pd,
locations="country",
locationmode='country names',
color="sum(confirmed)",
hover_name="country",
range_color=[1, 1000000],
color_continuous_scale='portland',
title='European countries with confirmed cases',
scope='europe',
height=800)
fig.show()
fig = px.bar(dataframe_pd.sort_values('sum(confirmed)',
ascending=False)[:10][::-1],
x='sum(confirmed)',
y='country',
color_discrete_sequence=[color_scheme.color_400],
title='Confirmed cases in Europe (top-10 countries)',
text='sum(confirmed)',
orientation='h')
fig.show()
return None
def visualize_confirmed_cases_countries(dbutils):
df_serbia_grouped = load_dataset(sys.argv[1], "confirmed_cases_serbia",
dbutils)
df_china_grouped = load_dataset(sys.argv[1], "confirmed_cases_china",
dbutils)
df_italy_grouped = load_dataset(sys.argv[1], "confirmed_cases_italy",
dbutils)
df_norway_grouped = load_dataset(sys.argv[1], "confirmed_cases_norway",
dbutils)
fig = go.Figure()
fig.add_trace(
go.Scatter(x=df_serbia_grouped['date'],
y=df_serbia_grouped['sum(confirmed)'],
name="Serbia",
line_color=color_scheme.color_500,
opacity=0.8))
fig.add_trace(
go.Scatter(x=df_china_grouped['date'],
y=df_china_grouped['sum(confirmed)'],
name="China",
line_color=color_scheme.color_700,
opacity=0.8))
fig.add_trace(
go.Scatter(x=df_italy_grouped['date'],
y=df_italy_grouped['sum(confirmed)'],
name="Italy",
line_color=color_scheme.color_900,
opacity=0.8))
fig.add_trace(
go.Scatter(x=df_norway_grouped['date'],
y=df_norway_grouped['sum(confirmed)'],
name="Norway",
line_color=color_scheme.color_300,
opacity=0.8))
fig.update_layout(
title_text="Overview of case growth in Serbia, China, Italy and Norway"
)
fig.show()
return None
def visualize_red_blood_cells_values(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "red_blood_cells_values", dbutils)
fig = px.histogram(dataframe_pd,
x="Red blood Cells",
title="Red blood Cells distribution",
color_discrete_sequence=[color_scheme.color_300],
opacity=0.8,
marginal="rug")
fig.show()
return None
def visualize_aggregate_age_result(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "aggregate_age_result", dbutils)
fig = px.line(dataframe_pd,
x="result",
y="avg(age)",
title="Average age/result distribution",
log_y=True,
color_discrete_sequence=[color_scheme.color_400])
fig.show()
return None
def visualize_hemoglobin_values(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "hemoglobin_values", dbutils)
fig = px.histogram(dataframe_pd,
x="Hemoglobin",
title="Hemoglobin distribution",
color_discrete_sequence=[color_scheme.color_500],
opacity=0.8,
marginal="rug")
fig.show()
return None
def visualize_confirmed_cases_recovery_rates(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "confirmed_cases_recovery_rates",
dbutils)
fig = px.bar(dataframe_pd,
x='recoveryRate',
y='country',
title='Recoveries per 100 confirmed cases (top-10)',
text='recoveryRate',
height=800,
orientation='h',
color_discrete_sequence=[color_scheme.color_500])
fig.show()
return None
def visualize_predictions_test_result_distribution(dbutils):
dataframe_pd = load_dataset(sys.argv[1],
"predictions_test_result_distribution",
dbutils)
fig = px.pie(dataframe_pd,
values='count',
names='result',
title="Statistics of test result distribution",
color_discrete_sequence=[
color_scheme.color_100, color_scheme.color_400
])
fig.show()
return None
def visualize_predictions_missing_values(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "predictions_missing_values",
dbutils)
dataframe_pd = dataframe_pd.rename(index={
0: 'count'
}).T.sort_values("count", ascending=False)
fig = px.bar(
dataframe_pd,
y="count",
color_discrete_sequence=[
color_scheme.color_400, color_scheme.color_500
],
title="Statistics of missing (null/nan) values across columns")
fig.show()
return None
def visualize_predictions(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "predictions", dbutils)
fig = go.Figure(data=[
go.Bar(y=dataframe_pd['value'],
x=[
'Random Forest classifier Accuracy',
'Decision Tree Accuracy', 'Logistic Regression Accuracy',
'Gradient-boosted Trees Accuracy'
])
])
fig.update_traces(marker_color=color_scheme.color_200,
marker_line_color=color_scheme.color_600,
marker_line_width=1.5,
opacity=0.6)
fig.update_layout(title_text='Comparison of classifier accuracy reports')
fig.show()
return None
def visualize_confirmed_cases_and_deaths_globally(dbutils):
dataframe_pd = load_dataset(sys.argv[1],
"confirmed_cases_and_deaths_globally", dbutils)
fig = px.line(dataframe_pd,
x="date",
y="sum(confirmed)",
title="Confirmed cases over time (logarithmic)",
log_y=True,
color_discrete_sequence=[color_scheme.color_400])
fig.show()
fig = px.line(dataframe_pd,
x="date",
y="sum(deaths)",
title="Death cases over time (logarithmic)",
log_y=True,
color_discrete_sequence=[color_scheme.color_900])
fig.show()
return None
def visualize_confirmed_cases_comparison(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "confirmed_cases_comparison",
dbutils)
df_melted = dataframe_pd.melt(
id_vars="date",
value_vars=['sum(recovered)', 'sum(deaths)', 'sum(active)'],
var_name='case',
value_name='count')
fig = px.area(df_melted,
x="date",
y="count",
color='case',
title='Cases over time',
color_discrete_sequence=[
color_scheme.color_200, color_scheme.color_400,
color_scheme.color_800
])
fig.show()
return None
def visualize_abstracts_words(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "paper_abstracts", dbutils)
text = dataframe_pd["clean_abstract"].values
stopwords = set(STOPWORDS)
fig = ff.create_distplot([dataframe_pd["sentiment_abstract"]],
["sentiment_abstract"],
colors=[color_scheme.color_400])
fig.show()
word_cloud = WordCloud(width=1000,
height=500,
stopwords=stopwords,
background_color="white",
max_words=25).generate(str(text))
fig = px.imshow(word_cloud.recolor(color_func=generate_custom_color,
random_state=3),
binary_compression_level=1,
title="Most commonly used words in abstracts")
fig.show()
return None
def visualize_care_relations(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "care_relations", dbutils)
fig = px.bar(dataframe_pd,
y="result",
x="Patient addmited to regular ward (1=yes, 0=no)",
color_discrete_sequence=[
color_scheme.color_400, color_scheme.color_500
],
title="Positive patients admitted to regular care")
fig.show()
fig_intensive = px.bar(
dataframe_pd,
y="result",
x="Patient addmited to intensive care unit (1=yes, 0=no)",
color_discrete_sequence=[
color_scheme.color_900, color_scheme.color_500
],
title="Positive patients admitted to intensive care")
fig_intensive.show()
return None
def visualize_predictions_value_distribution(dbutils):
dataframe_pd = load_dataset(sys.argv[1], "predictions_value_distribution",
dbutils)
fig = make_subplots(
rows=3,
cols=3,
subplot_titles=("Hemoglobin/Exam Result", "Platelets/Exam Result",
"Eosinophils/Exam Result",
"Red blood Cells/Exam Result",
"Lymphocytes/Exam Result", "Leukocytes/Exam Result",
"Basophils/Exam Result", "Monocytes/Exam Result",
"Hematocrit/Exam Result"))
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Hemoglobin"],
mode='markers',
marker=dict(color=color_scheme.color_900)),
row=1,
col=1)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Platelets"],
mode='markers',
marker=dict(color=color_scheme.color_800)),
row=1,
col=2)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Eosinophils"],
mode='markers',
marker=dict(color=color_scheme.color_700)),
row=1,
col=3)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Red blood Cells"],
mode='markers',
marker=dict(color=color_scheme.color_600)),
row=2,
col=1)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Lymphocytes"],
mode='markers',
marker=dict(color=color_scheme.color_500)),
row=2,
col=2)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Leukocytes"],
mode='markers',
marker=dict(color=color_scheme.color_400)),
row=2,
col=3)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Basophils"],
mode='markers',
marker=dict(color=color_scheme.color_300)),
row=3,
col=1)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Monocytes"],
mode='markers',
marker=dict(color=color_scheme.color_200)),
row=3,
col=2)
fig.add_trace(go.Scatter(x=dataframe_pd['SARS-Cov-2 exam result'],
y=dataframe_pd["Hematocrit"],
mode='markers',
marker=dict(color=color_scheme.color_100)),
row=3,
col=3)
fig.show()
return None
